machine-learning.data-sets: renaming slots in the data-set tuple

"features" is a better name for the slot than "data"

diff --git a/extra/machine-learning/data-sets/data-sets.factor b/extra/machine-learning/data-sets/data-sets.factor
index 018c4856c98c912712f81cf9c231e15ede9fb2d0..34c73616916b8d7c1de059ca4925eba002eb2c0f 100644 (file)
--- a/extra/machine-learning/data-sets/data-sets.factor
+++ b/extra/machine-learning/data-sets/data-sets.factor
@@ -6,8 +6,10 @@ kernel math.parser sequences splitting ;
 
 IN: machine-learning.data-sets
 
-TUPLE: data-set data target target-names description
-    feature-names ;
+TUPLE: data-set
+    features targets
+    feature-names target-names
+    description ;
 
 C: <data-set> data-set
 
@@ -37,23 +39,25 @@ PRIVATE>
     ! Omits the identifiers which are not so interesting.
     [ but-last [ string>number ] map ] map
     [ [ rest ] map ] [ [ first ] map ] bi
+    { "a1" "a2" "a3" "a4" "a5" "a6" }
     { "no" "yes" }
     "monks.names" load-file
-    { "a1" "a2" "a3" "a4" "a5" "a6" } <data-set> ;
+    <data-set> ;
 
 : load-iris ( -- data-set )
     "iris.csv" load-table-csv
     [ [ unclip-last ] { } map>assoc unzip ] [ 2 tail ] bi*
-    "iris.rst" load-file
     {
         "sepal length (cm)" "sepal width (cm)"
         "petal length (cm)" "petal width (cm)"
-    } <data-set> ;
+    } swap
+    "iris.rst" load-file
+    <data-set> ;
 
 : load-linnerud ( -- data-set )
     data-set new
         "linnerud_exercise.csv" load-table
-        [ >>data ] [ >>feature-names ] bi*
+        [ >>features ] [ >>feature-names ] bi*
         "linnerud_physiological.csv" load-table
-        [ >>target ] [ >>target-names ] bi*
+        [ >>targets ] [ >>target-names ] bi*
         "linnerud.rst" load-file >>description ;

diff --git a/extra/machine-learning/decision-trees/decision-trees.factor b/extra/machine-learning/decision-trees/decision-trees.factor
index b843e7847f09202827f66bccd1d9b3c21bf56a7f..4365582e99f7f61d0ee96583dd41e272cc93b28b 100644 (file)
--- a/extra/machine-learning/decision-trees/decision-trees.factor
+++ b/extra/machine-learning/decision-trees/decision-trees.factor
@@ -20,12 +20,12 @@ IN: machine-learning.decision-trees
     [ [ entropy2 ] [ length ] bi * ] map-sum ; inline
 
 :: average-gain ( dataset idx -- gain )
-    dataset target>> :> target
-    dataset data>> :> data
-    data target zip :> data-target
-    data-target idx subsets-weighted-entropy :> weighted
+    dataset targets>> :> targets
+    dataset features>> :> features
+    features targets zip :> features-targets
+    features-targets idx subsets-weighted-entropy :> weighted
 
-    target entropy2 weighted data length / - ;
+    targets entropy2 weighted features length / - ;
 
 : highest-gain-index ( dataset -- idx )
     dup feature-names>> length <iota> [